PROJECT SUMMARY We previously identified tumor cell PD-L1 membranous (cell surface) expression in pretreatment biopsies as a correlate of the likelihood of response to anti-PD-1 therapy. This general finding of an association of PD-L1 expression with tumor response to anti-PD-1 therapy has now been substantiated across thousands of patients with certain tumor types treated with both anti-PD-1 and anti-PD-L1. However, while PD-L1 expression enriches for response to anti-PD-1/PD-L1, it is not sufficient. Other related features which have been suggested to improve on the PD-L1 biomarker includes whether PD-L1 is expressed on a tumor cell or immune cell, the CD8 cell density at the tumor's leading edge, CD8:FoxP3+ cell ratio, the distance between PD-1 and PD-L1 protein expression, and PD-1 expression levels (low vs. high). We have developed a quantitative multiplex immunofluorescence assay which captures all of these features and includes PD-L1, PD-1, CD8, FoxP3, CD68, a tumor marker (e.g. Sox10/S100 for melanoma), and DAPI. Through this assay, it is possible to define which cell type is expressing PD-1 and PD-L1 and to enumerate specific cellular subsets. It is also possible to include spatial parameters, including the distance between PD-1 and PD-L1, which have not previously been included in predictive or prognostic surgical pathology specimen-based assays. As a part of the parent R01, we have shown that our assay has increased sensitivity and specificity for response when compared to the assessment of PD-L1 expression alone in multiple tumor types, including melanoma, non-small cell lung carcinoma, Merkel cell carcinoma, head and neck squamous cell carcinoma, amongst others. We have completed formal assay validation studies at a single site, and in some instances have completed comparisons between two sites. The purpose of this proposal is to perform inter-site validation of the assay amongst four major academic sites (Johns Hopkins, Yale University, MD Anderson, Providence Portland Medical Center). This proposal includes two main Aims: 1) validation of the staining and scoring reproducibility across all four academic sites on tonsil and archival melanoma specimens, and 2) the establishment of final assay thresholds using ~100 pre- treatment specimens from patients with melanoma treated with anti-PD-1. For Aim 2, staining for each case will also be performed across three academic sites to provide additional information regarding assay reproducibility. Discovery and Validation cohorts will be used to establish final assay parameters linked to clinical outcomes. The deliverable of the study is a refined, multiplex biomarker assay for response/resistance to anti-PD-1 that has been validated across multiple academic sites. The result will be a multiplex IF assay that is suitably staged for advanced development aimed at clinical implementation. Such an assay will facilitate more precise therapeutic guidance in patients receiving anti-PD-1. While melanoma is the focus of the current grant proposal, our preliminary results suggest that this assay will also have great value in numerous other solid tumor types.